002-Akbar 2012 Performance Measurement in Indonesia-The Case of Local Government

Pacific Accounting Review Performance measurement in Indonesia: the case of local government Rusdi Akbar Robyn Pilcher Brian Perrin

Article information:

Downloaded by Universitas Gadjah Mada At 09:27 11 September 2015 (PT)

To cite this document: Rusdi Akbar Robyn Pilcher Brian Perrin, (2012),"Performance measurement in Indonesia: the case of local government", Pacific Accounting Review, Vol. 24 Iss 3 pp. 262 - 291 Permanent link to this document: http://dx.doi.org/10.1108/01140581211283878 Downloaded on: 11 September 2015, At: 09:27 (PT) References: this document contains references to 87 other documents. To copy this document: [email protected] The fulltext of this document has been downloaded 1274 times since 2012*

Users who downloaded this article also downloaded: Rusdi Akbar, Robyn Ann Pilcher, Brian Perrin, (2015),"Implementing performance measurement systems: Indonesian local government under pressure", Qualitative Research in Accounting & Management, Vol. 12 Iss 1 pp. 3-33 http://dx.doi.org/10.1108/QRAM-03-2013-0013 Louise Kloot, (1999),"Performance measurement and accountability in Victorian local government", International Journal of Public Sector Management, Vol. 12 Iss 7 pp. 565-584 http:// dx.doi.org/10.1108/09513559910308039 Harun Harun, Karen Van Peursem, Ian Eggleton, (2012),"Institutionalization of accrual accounting in the Indonesian public sector", Journal of Accounting & Organizational Change, Vol. 8 Iss 3 pp. 257-285 http://dx.doi.org/10.1108/18325911211258308

Access to this document was granted through an Emerald subscription provided by emerald-srm:273599 []

For Authors If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service information about how to choose which publication to write for and submission guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.

About Emerald www.emeraldinsight.com Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online products and additional customer resources and services. Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.


*Related content and download information correct at time of download.

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0114-0582.htm

PAR 24,3

Performance measurement in Indonesia: the case of local government

262

Rusdi Akbar Fakultas Ekonomi, Universitas Gadjah Mada, Yogyakarta, Indonesia, and

Robyn Pilcher and Brian Perrin Downloaded by Universitas Gadjah Mada At 09:27 11 September 2015 (PT)

School of Accounting, Curtin University, Perth, Australia Abstract Purpose – The purpose of this paper is to investigate the implementation of performance measurement systems (PMSs) in Indonesian local government (ILG) using Smart PLS. Couched within an institutional theory framework, it explores a conceptual model developed to explain the hypothesised relationships between technical and organisational factors and the development and use of performance indicators and accountability practices. Design/methodology/approach – Surveys were sent to senior finance officers in all local governments (457) across Indonesia with a response rate of 21.4 percent being achieved. Smart PLS was used to assess the quality of the data and analyse the research model proposed. Findings – Findings revealed that ILGs developed performance indicators more to fulfil regulatory requirements than to make their organisation more effective and efficient. As a way of increasing the success of PMS implementation management commitment through good leadership was found to be a major contributor. Coercive pressure from central government impacted on the result as did normative isomorphism by way of widespread training by universities (and others) and the subsequent sharing of this knowledge. Practical implications – The findings will assist Indonesian central government formulate future government policy as well as design appropriate strategies for implementing the second wave of (bureaucratic) reform. Originality/value – Set in a local government environment in a developing country, this research is original and makes three major contributions. First, it provides an understanding of factors influencing the development and use of performance measures in the ILG context. Second, the use of Smart PLS is original in this context and fills a gap in the literature examining local government PMS. Last, the existence of institutional isomorphism reaffirms that this theory is still applicable in the twenty-first century and relevant as an explanator of the results in the context examined here. Keywords Performance measurement systems, Indonesia, Local government, Institutional theory, Isomorphism, PLS analysis, Performance management Paper type Research paper

1. Introduction Many government entities in developed countries have introduced elements of new public management (NPM)[1] (Hood, 1991, 1995; ter Bogt, 2004) which is based on a fundamental concept that public sector organizations can, and even should, borrow Pacific Accounting Review Vol. 24 No. 3, 2012 pp. 262-291 q Emerald Group Publishing Limited 0114-0582 DOI 10.1108/01140581211283878

The authors would like to acknowledge the anonymous reviewers at the AFAANZ 2010 Conference for their feedback on an earlier version of the paper. The authors would also like to thank the two anonymous reviewers who provided feedback as part of this journal’s review process.


many management strategies from the private sector. In Indonesia the government introduced performance measurement reporting to respond to public demand on productivity, transparency and accountability. According to Cheung (2011, p. 131), the NPM and good governance models were the two “dominant paradigms” having the “greatest impact on Asian (in which he includes Indonesia) institutional reforms”. This study focuses on two pivotal aspects in the public sector – accountability and performance measurement. Over the last 40 years the idea of performance measurement has been the focus of considerable attention from both academics and practitioners (Neely, 1999; Kihn, 2010). Initially, the research in the field mainly concentrated on performance measurement in the private sector (Kaplan, 1983; Johnson and Kaplan, 1987; Chenhall and Smith, 2007). However, since then there have been a number of studies addressing the issues of performance measurement in the public sector (Hood et al., 1998; Cavalluzzo and Ittner, 2004; Micheli and Neely, 2010). In Indonesia, the interest in performance measurement began to emerge at the beginning of the reform era in 1999 when President B.J. Habibie signed a presidential instruction (Inpres No. 7/1999), the Performance Accountability Report of State Apparatus (known as LAKIP). Coinciding with the first wave of public sector reform, this research investigates the implementation of performance measurement systems (PMS) in Indonesian local government (ILG). A second generation structural equation modelling technique known as partial least squares (PLS) was used to analyse the proposed model and relationships. Based on a survey of ILGs, the research explores a conceptual model adapted and developed to explain the hypothesised relationships between technical and organisational factors and the development and use of performance indicators and accountability practices. The theoretical development and interpretation of this research is drawn from institutional theory. A major contribution of this research is that it provides an understanding of factors influencing the development and use of performance measures which, in turn, could be used to formulate future government policy. Further, it fills a gap in the literature exploring PMS and accountability in a developing country – in this case, Indonesia. This paper begins with a brief literature review followed by an explanation of the theoretical framework employed. Section 4 then considers the research model and hypotheses development, and Section 5 discusses the research method used. Section 6 presents the results of the research using SmartPLS, with these results discussed in the penultimate section. A conclusion and suggestions for future research ends this paper. 2. Literature review NPM and accountability NPM was introduced in Europe in the 1980s. Lapsley and Pallot (2000) considered this framework separately from that of institutional theory, whereas Bovaird and Downe (2006) use institutional isomorphism to understand NPM-type reforms. This paper adopts the latter framework as it considers whether, with the introduction of PMS, institutional isomorphism – particularly coercive – plays a role in the implementation processes adopted by ILG. The development of NPM was seen as a means by which to enhance accountability and transparency of governments and this, in turn, required performance information that was more comparable, relevant and useful for decision-making within the public sector. Romzek and Dubnick (1998, p. 6), define accountability as “a relationship in which

Performance measurement in Indonesia 263

PAR 24,3


264

an individual or an agency is held to answer for performance that involves some delegation of authority to act.” Government organisations are created by the public, for the public, and need to be accountable to the public. Given the myriad definitions and variations on the term “accountability”, for the purposes of this paper, accountability is separated into two components – internal accountability (the accountability of management to the major/head of district (elected official)), and external accountability (accountability of ILG to its stakeholders such as central government, parliaments and citizens). Performance measurement/PMS Managing and measuring performance has been one of the key drivers in the reform of the public sector (Gianakis, 2002). Osborne and Gaebler (1992) claimed that to win public support, government officials needed to demonstrate the results achieved. From this, one could ascertain that local government should manage and measure its performance using PMS in order to better serve the citizens and in turn gain their support. As the form of government that is closest to the people, ILG has been recognised as the sphere of government that has the largest capacity to deliver real outcomes at a local level (LOGOTRI, 2003). This development, in turn, should increase the quality of life for its citizens. In order to do this, however, local government must have some way of measuring the success, failure, and progress achieved in the pursuit of the objectives. Increasingly, performance measurement is considered to be a vital tool for local government to not only measure its activities, but also to then provide feedback into the management process to help improve future performance (LOGOTRI, 2003). Further, the development and use of performance measures is considered a key element of NPM implementation (Lapsley, 2008; Boston, 2011). Research on performance measurement in the Indonesian public sector is quite scarce. One study by Riandi (2003) found there was still an expectation gap between local government authorities and the community about the public service delivery. This gap occurred due to the lack of relevant performance indicators available to distinguish success from failure in public services delivery (Mahsun, 2005). McLeod (2005, p. 1) examined the reform process in Indonesia as a whole and provided some explanations from the private sector for what he saw as “demonstrably poor performance” in Indonesia. Adoption of PMS as an important component of management reform in the public sector has become more prevalent in the last two decades. Some scholars claimed that the arguments for this movement were mainly rational or technical in nature, and that PMS was adopted as an effort towards achieving an efficient and effective organisation (Andrews et al., 2006; Meier et al., 2006). As a result of PMS implementation it was envisaged that better substantive performance would occur due to the expected benefits of enhanced efficiency, accountability, and quality of public service delivery. For example, Andrews et al. (2006) argued that PMS be included in both theoretical and empirical models of public sector organisational performance. However, a different point of view is offered by institutional theory, which argues that the main reason underlying organisational change is gaining legitimacy rather than improving substantive performance (Ashworth et al., 2009). Reasons for the shift in characteristics, therefore, were not only due to rational or technical reasons, but often political or legitimising ones (Ashworth et al., 2009; Scott, 1987).


3. Theoretical framework Introduction As a theoretical framework, institutional theory is a very broad concept and has many strands that have evolved since the development of the theory in the early twentieth century. The foundation of early institutionalism (old institutionalism) and its application to organisations has been outlined by Selznick (1949, 1996). In contrast, modern institutionalism (new institutionalism) emerged in the 1970s, as has been shown in the work of Silverman (1971) followed by Meyer and Rowan (1977), Zucker (1977) and DiMaggio and Powell (1983) who extended Meyer and Rowan’s (1977) concept of isomorphism to encompass the organisational field. Selznick (1996) compares the “old” with the “new” in an attempt to explain the differences – such as the definition of institutionalisation – as well as promote the concept that institutional theory should be able to embrace all the classifications and definitions embedded in both the “old” and “new”. One major difference between the two is the view of the “institution”, with researchers like DiMaggio and Powell (1991) espousing isomorphism and then the doctrine of “power, subordination and responsibility” (Selznick, 1996, p. 272). More recently, academics have considered institutional theory in varying forms: Modell (2004, 2005) and Tsamenyi et al. (2006) – utilise the new institutional sociology to explain organisational change; Brignall and Modell (2000) – who consider that institutional theory could explore organisational change with the addition of the interests and power from various stakeholders; Thornton and Ocasio (2008) – who contend that institutional logics provide a bridge between some of the earlier institutional theories; Lounsbury (2008) – offers alternatives to isomorphism; Ashworth et al. (2009) – shows the benefits of using isomorphism as an explanator of change; and Collin et al. (2009) – who suggest integrating positive accounting theory with institutional theory. It is obvious from this brief summary that institutional theory has many dimensions. In the field of organizational study, the concepts of institution and institutionalization have been defined in many different ways. One definition proposed by Scott (1987) was institutionalisation as a process of creating reality. This definition is relevant here as this form of institutional theory is based on a shared social reality which, in turn, creates a human construction in social interaction. In other words, organisations (councils) operating in the same field (local government) react similarly and, over time, develop similar responses or adopt similar practices when facing certain situations. DiMaggio and Powell (1983, p. 147) explain: “Institutionalization is a rational response; one of the key outcomes of institutionalization is a homogeneous organizational structure.” Debate around nuances associated with variations on the “old” and then “new” institutional theory continue today (Falkman and Tagesson, 2008; Lounsbury, 2008; Ashworth et al., 2009). Although many of these nuances would have fitted with the research objective here (for example, that proposed by Schneiberg and Soule (2005) which considered political and cultural aspects of the diffusion process), the more traditional form of institutional isomorphism was used to inform the instrument and its subsequent analysis. As a developing country, in many ways Indonesia is still in its infancy in regards to the development of PMS – particularly during the first wave of reforms. Coercive and normative pressures played a role in the adoption process as will be explained in the discussion section below.


PAR 24,3


266

Isomorphism DiMaggio and Powell (1983) proposed that over time, in a well-established field, organisations tend to move towards homogenisation, even though they show considerable diversity at first. The term used that best describes the process of homogenisation is isomorphism. Although there are said to be different types of isomorphism (Meyer and Rowan, 1977; DiMaggio and Powell, 1983), institutional isomorphism is used here. It is a useful concept in an environment where “politics and ceremony” (DiMaggio and Powell, 1983, p. 150) are embedded in organizational life. Meyer and Rowan (1977, p. 348) argue that isomorphism had three consequences for organisations: (a) they incorporate elements which are legitimated externally, rather than in terms of efficiency; (b) they employ external or ceremonial assessment criteria to define the value of structural elements; and (c) dependence on externally fixed institutions reduces turbulence and maintains stability.

DiMaggio and Powell (1983) identified three isomorphic forces – coercive, mimetic and normative. Given this is not a longitudinal study, the first and latter isomorphic pressures are considered for this project. Coercive isomorphism stems from political influence and the pursuit of legitimacy. This pressure comes from both formal and informal pressures from other organizations. Normative isomorphism, on the other hand, is usually associated with professionalism and the way both formal and informal collaboration can lead to more homogenisation of organisations. Despite myriad views on other aspects of institutional theory and questions regarding its applicability, institutional isomorphism is still being used successfully worldwide (see for example, Tuttle and Dillard (2007) – USA; Kim et al. (2009) – Korea; Arnaboldi et al. (2010) – Italy). Performance measurement and isomorphism Over the years management control systems with performance measurement playing an integral role, have been studied from functionalist, behavioural, interpretive and critical perspectives. Recent studies, however, have been particularly influenced by institutional theories (Berry et al., 2009). Studies that adopted this theory assumed that organizations compete not only for resources and customers but also for political power and institutional legitimacy. Institutional theory suggests that organizations pursue “legitimacy by conforming to isomorphic pressures in their environment” (Ashworth et al., 2009, p. 1). Therefore, from this perspective, the logics of change in PMS are institutionalized into organizations by means of three processes: coercive, mimetic, and normative (DiMaggio and Powell, 1983). This study investigates the implementation of PMS in ILG and considers senior finance officers’ perceptions of performance measurement within an institutional theory framework. The way performance measurement is used within government may depend on the power relationship between its constituents and itself (Pilcher and Dean, 2009). Demonstrating this, Pilcher (2011, p. 368) examined local government’s reporting under a performance measurement regime designed to ensure councils “performed or perished”. Brignall and Modell (2000, p. 282) also examined the implementation of PMS and claimed that to extend previous research in the area, it was necessary to “shift the attention to the power and pressures exerted by different groups of stakeholders and how these affect the use of performance information in organisations.”


Hence, in a decentralised government, such as Indonesia, the central authority normally has greater coercive power over local government than other constituents. In Indonesia, the sources of isomorphic pressures potentially come from central government via enactment of laws and regulations that will affect local government. These regulations include Inpres No. 7/1999 that requires all government entities, including local government, to report on performance to central government. This coercive pressure tends to occur due to most local governments being heavily dependent on central government for their financial resources or recognition via some award system. Even though ILGs are required to submit performance reports to central government, this does not mean that they actually use performance information in their day-to-day management practices (as found by various authors, including, for example, in a US study by de Lancer Julnes and Holzer (2001)). Therefore, an understanding of factors influencing the development and use of performance measures is important. The knowledge of these factors could be utilized to evaluate and improve future government policy. Although policy adoption has been said to rely on “factors from rational and technocratic theory” (de Lancer Julnes and Holzer, 2001, p. 693), actual implementation is more complex and is influenced by “political and cultural considerations” (de Lancer Julnes and Holzer, 2001, p. 693). Hence, this study contributes to the literature and understanding by government policymakers by considering factors associated with development and use of performance measures in Indonesia. The third of DiMaggio and Powell’s (1983) isomorphic pressures, but the second to be used here, is normative isomorphism. This element of pressure is normally developed by professional and occupational groups (Rahaman et al., 2004). Given the low level of human resources capacity in ILGs, there has been a trend in the last decade to give more attention to the education of government employees and managers. As the demand emerged, many universities in Indonesia have offered programmes (degree and non-degree) that are specially designed to respond to government employees and managers’ needs. DiMaggio and Powell (1983) argued that the more educated the workforce becomes, in terms of academic qualification and participation in professional and trade associations, the greater the extent to which the organization becomes similar to other organizations in the field. 4. Research model and hypotheses formulation The objective of this research is to develop a conceptual model that identifies the factors that influence the development and use of performance indicators in ILG. The primary research question is: RQ1. What factors affect the development and use of performance indicators and accountability practices in ILG? The issues identified in previous studies were used to formulate the theoretical framework described in Section 3 and secondary research questions that drive the research: RQ2. Do metric difficulties, technical knowledge, management commitment, and legislative requirements influence the development and use of performance indicators in ILG?


PAR 24,3


268

RQ3. Does institutional isomorphism exist in the development and use of PMS and accountability practices in ILG? The conceptual model outlined in this paper combines components of the accountability model of Wang (2002) with the model of performance measurement from Cavalluzzo and Ittner (2004). It is modified and extended here to not only fit within the ILG context but also to be analysed within an institutional theory framework. Cavalluzzo and Ittner’s (2004) study considered factors associated with performance measurement in the US Government. Questions related to the first four dependent variables (described and tabulated in Appendix 1) were adapted from their study with one major difference being the modification of five items (1, 2, 3, 5, and 6 under development of indicators in Appendix 2) to include a family of measures derived from the Indonesian LAKIP guidelines. In regards to the final dependent variable – external accountability – Wang’s (2002) model (in itself a permutation of items from other sources) was used as a template and the eight items from the performance accountability construct were adjusted to suit the ILG context. In regards to the five independent variables, questions were again adapted from Cavalluzzo and Ittner (2004) for the first four variables whilst four items were taken from Wang (2002) to measure the level of organisational capacity (the final independent variable). Again, differences existed with the first relating to “technical knowledge” which, although based on the “training” construct in Cavalluzzo and Ittner (2004), differs in the areas it addresses so that it fits with the ILG environment (and, again, LAKIP). Legislative requirements are also based on LAKIP. Combining components of each of these two studies not only allowed for a comprehensive examination of factors affecting the development and use of performance indicators, but also provided an important aspect for ILG – particularly within NPM – being accountability. Hence, all three research questions detailed above could be addressed. Figure 1 shows the conceptual model tested in this analysis and Appendix 1 contains a copy of the constructs used in the model. The hypothesised relationships between the independent and dependent variables are discussed in the following sections along with the hypotheses proposed. Dependent variables Development and use of performance indicators. Various studies regarding the use of performance measurement conducted in local government suggest that the practice of collecting performance indicators, at least at a rudimentary level, is fairly well established in many countries, including the USA, the UK and Australia (Pilcher and Dean, 2009). Despite this growing practice, previous studies found that the development and use of performance indicators in the public sector is problematical with very few local governments using performance indicators (Bellamy and Kluvers, 1995; Ammons, 1995). As shown in Figure 1, the model includes five dependent variables: (1) development of performance indicators; (2) managerial use of indicators; (3) higher use of indicators; (4) internal accountability; and (5) external accountability.

Performance measurement in Indonesia

Metric Difficulties

Development of Indicator

Technical Knowledge

269


Managerial Use of Indicator Internal Accountability

Management Commitment Higher Use of Indicator

External Accountability

Legislative Mandates

Organisational Capacity

Notes:

P < 0.001 and P < 0.01;

P < 0.05;

no association

In this study, development, accountability and use of performance indicators are assessed by asking respondents to answer a series of survey questions that refer to the development and adoption of different types of performance measures used by the organisation. The first construct reflects the extent of performance indicators developed by ILG organisations. Managerial use of indicators refers to the extent to which performance indicators are used by mid-level managers to manage the organisation. Higher use of indicators refers to measures that are used by top-level management to make funding decisions and manage high-level changes. Accountability in local government is measured by financial and non-financial performance indicators. The internal accountability construct determines the extent of the accountability relationship between the supervisor and subordinate (for example, in ILG between the general manager and the council members), whereas the external accountability construct measures the extent of the relationship between ILG organisations and their stakeholders (refer to Cheng (1994), for an example of the myriad stakeholders associated with local government).

Figure 1. Conceptual model

PAR 24,3


270

Independent variables There were five independent variables identified within the model (Figure 1): (1) metric difficulties; (2) technical knowledge; (3) management commitment; (4) legislative requirement; and (5) organisational capacity. Metric difficulties. The first independent variable refers to measurability of organizational output or outcome. In other words, it considers the ability to “define and assess metrics that capture desired actions and outcomes” (Cavalluzzo and Ittner, 2004, p. 247). In public sector organizations the complexity of programmes is not always the same. Some outputs and effects are relatively easy to measure yet some are more difficult. For example, in Indonesia, Mahmudi (2003) found that local government managers had difficulties in determining higher level indicators such as outcome, benefit and impact, therefore in practice they placed more weight on input measures rather than outcome measures. Given the situation in Indonesia it is logical to expect that there will be a tendency in performance measurement users to give more weight to the easy-to-measure indicators: H1a. Development of performance indicators is negatively associated with metric difficulties. H1b. Managerial use of performance indicators is negatively associated with metric difficulties. H1c. Higher use of performance indicators is negatively associated with metric difficulties. Technical knowledge. An organizational factor that is expected to influence the development and use of performance indicators is the extent to which training on related knowledge is provided to support the implementation (Shields, 1995; Cavalluzzo and Ittner, 2004). According to Sukarno (2006), a lack of understanding of systems designed to support the implementation of performance indicators has impacted on their development. Technical knowledge enables improvement in the ability of internal stakeholders to understand and use PMS, and should positively improve the development and use of performance indicators (de Lancer Julnes and Holzer, 2001; Laurensius and Halim, 2005). Different types of efforts, from technical training to formal degree programmes, have been undertaken in Indonesia to increase the knowledge of government employees and officers. From this perspective, normative mechanisms as suggested by DiMaggio and Powell (1983) may also take place in practice: H2a. Development of performance indicators is positively associated with related technical knowledge. H2b. Managerial use of performance indicators is positively associated with related technical knowledge. H2c. Higher use of performance indicators is positively associated with related technical knowledge.


Management commitment. Implementation of PMS in government requires changes in the operation, personnel, structure, or even culture of the organization. It is important to build high levels of commitment amount senior management first, followed by support from middle managers and staff (Fernandez and Rainey, 2006). de Lancer Julnes and Holzer (2001) found that internal stakeholder support was positively related to the adoption of PMS. As well, management also provides the political support needed to encourage and motivate individuals who resist the innovation. In addition to financial resources, time and personnel, the existence of internal commitment, especially by top-level management within an organization, is required for the successful implementation of performance reporting in Indonesia (Sukarno, 2006): H3a. Development of performance indicators is positively associated with management commitment. H3b. Managerial use of performance indicators is positively associated with management commitment. H3c. Higher use of performance indicators is positively associated with management commitment. H3d. Internal accountability is positively associated with management commitment. H3e. External accountability is positively associated with management commitment. Legislative mandates. Institutional theory suggests that legislative mandates (or regulation requirements) is an organizational factor that is relevant to the success of reform implementation in government organizations (Brignall and Modell, 2000). Furthermore, in an institutional environment such as ILG which primarily depends on an external organization (i.e. central government) for its financial support, external bodies have the authority to impose organizational practices on subordinate units. Consequently, subordinate organizations will implement the required practices, but the actual results tend to be superficial (Scott, 1987). The implementation process of Inpres No. 7/1999 on accountability in Indonesia was highly centralistic. Cavalluzzo and Ittner (2004) found that implementation of externally mandated PMS was merely executed to meet legal requirements. This resulted in the measures having little influence on internal operations. In other words, it was just more “to conform than to perform” (Barreto and Baden-Fuller, 2006, p. 1559). From this perspective, coercive mechanisms as suggested by DiMaggio and Powell (1983) may take place in practice: H4a. Development of performance indicators is positively associated with legislation requirements. H4b. Managerial use of performance indicators is positively associated with legislation requirements. H4c. Higher use of performance indicators is positively associated with legislation requirements. H4d. Internal accountability is positively associated with legislation requirements. H4e. External accountability is positively associated with legislation requirements.


PAR 24,3


272

Organizational capacity. The final factor employed considers the impact of organizational capacity/resources on accountability practices. Prior research found that insufficient resources were often a cause of implementation failure (Ammons and Rodrigues, 1986). The importance of capable staff in regard to the intensive use of performance measurement information has been studied by researchers such as Wang and Berman (2001). They found a positive relationship between staff competence and performance measures (Wang and Berman, 2001). Furthermore, van Dooren (2005) argued that lack of resources is an issue in performance measurement practice. Laurensius and Halim (2005) found that in Indonesia resources have a statistically significant effect on developing performance indicators. These findings confirmed those of de Lancer Julnes and Holzer (2001) who claimed that committed resources had a significant effect on the development of performance indicators: H5d. Internal accountability is positively associated with organizational capacity. H5e. External accountability is positively associated with organizational capacity. 5. Research method Table I provides a summary of the proposed model variables and the number of items used to measure each variable. The structural equation model offered here is divided into two key components; the measurement model and the structural model. The measurement model, or outer model, focuses on the relationship between the unobserved latent variables that cannot be directly measured and the measurable items known as the manifest variables (Hair et al., 2012; Chapman and Kihn, 2009). The structural model, also known as the inner model, focuses on the hypothesised relationships or paths between the latent variables (Hair et al., 2012). All of the measureable items used in this research are classified as reflective indicators. Survey In September 2008, surveys were sent to senior finance officers who are responsible for performance reporting in all local governments across Indonesia[2]. The constructs used to develop the questionnaire were adapted from surveys used in Western countries (Cavalluzzo and Ittner, 2004; Wang, 2002), hence all the questions needed to be translated from English into Bahasa Indonesia. There are several techniques available with the direct translation method, and a relatively simple and straightforward but effective

Table I. Research model variables

Latent variable

Short code

Manifest variables

Development of indicator Managerial use of indicator Higher use of indicator Internal accountability External accountability Metric difficulties Technical knowledge Management commitment Legislative mandate Organisational capacity

Dev Muse HUse IAcc EAcc Met Kno Com Leg Cap

Dev1 to Dev7 MUse1 to Muse8 HUse1 to HUse3 IAcc1 to IAcc4 EAcc1 to EAcc8 Met1 to Met5 Kno1 to Kno5 Com1 to Com3 Leg1 to Leg2 Cap1 to Cap4 Total

No. of items 7 8 3 4 8 5 5 3 2 4 49


method was used here (Usunier, 1998). A qualified and experienced translator from the National Accreditation Authority for Translators and Interpreters was invited to translate the questions and instructions used in the research instruments. To mitigate the disadvantage of direct translation, the study conducted a pilot test to ensure that a satisfactory level of reliability was achieved (Usunier, 1998). As of 2008, there were in total 457 local governments within 33 provinces with a total population of 230 million (Indonesian Bureau of Statistics, 2011). Of these, 211 were newly established as a result of the districts and cities separation process over the previous five years. Table II presents the distribution of responses from the survey. Due to invalid data, two responses were ineligible for further analysis and consequently only 98 usable responses (equivalent to a 21.4 per cent effective response rate) were analysed. Among the respondents, 44 per cent were from Java with 56 per cent from outer-Java. Most respondents (78 per cent) came from districts and only 22 per cent came from cities. When comparing the response rate of local government by type, the sample is fairly comparable with 21.5 per cent from districts and 23.2 per cent from cities. With regard to location, there was only a 16.4 per cent response rate from outer-Java, whilst that from Java was 38.3 per cent. The relatively low response rate can be explained by two factors. First, and the most likely reason, was the lack of experience in performance reporting of the 211 newly established local governments. This was supported by the fact that almost all responses were from local governments in existence long before the reform era (old local governments). Second, anecdotally, very low responses are normally expected from local governments in outer-Java. To ensure no response bias, the 14 late responses were compared to the 84 earlier responses using the Mann-Whitney test (Field, 2009). Analysis revealed that for all variables employed the mean rank between earlier and later responses was not significantly different. Demographic information of the respondents is summarized in Appendix 2. One point worth highlighting is that only 9 per cent of the respondents were female and 89 per cent were male (2 per cent missing). These figures are representative of the population of the study given the fact that the percentage of female senior finance officers in Indonesian civil servants is only 9.5 per cent.


6. PLS analysis and results SmartPLS 2.0 (Ringle et al., 2005) and SPSS 18 were used to assess the quality of the survey data and analyse the research model proposed. SmartPLS is a second Sent (457) Frequency Location In-Java Outer-Java Type Districts Cities

%

Received (100) Frequency %

Response rate (21.9%) (%)

115 342

25.2 74.8

44 56

44.0 56.0

38.3 16.4

362 95

79.2 20.8

78 22

78.0 22.0

21.5 23.2

Note: Due to invalid data, two responses were excluded and 98 were further analysed

Table II. Distribution of responses

PAR 24,3


274

generation structural equation modelling technique that has been used in a number of accounting studies (see Hall (2008), for a list of studies). It is particularly suited to this study because of the small sample size and complexity of the proposed model (Hair et al., 2012). Several authors, including Barclay et al. (1995), suggest the sample size should be no lower than ten times the number of items within the most complex construct of the model tested. Based on this rule, the sample size for this study would need to be a minimum 80 cases (n ¼ 8 items related to managerial use of indicator and external accountability by ten). Here there are 98 valid cases in this study satisfying the requirement for a sound PLS analysis. One limitation of this rule is that it does not consider factors such as size and reliability. Such factors are discussed shortly. What it does do is provide an estimate of the minimum sample size requirements and thus has been used by many for robust PLS estimations (Hair et al., 2012). Data examination Scanning of data on a line-by-line basis confirmed the absence of errors and the data set was ready for further SmartPLS procedures. As PLS regression analysis does not require the data to be normally distributed (Chin et al., 2003), tests for normality were not necessary. As well, goodness-of-fit measures were not appropriate here. Instead, similar to the process used in prior PLS research, fit was evaluated by the overall incidence of significant relationships between constructs and the explained variance of the dependent variables (Chapman and Kihn, 2009; Chenhall, 2005). Five exogenous (dependent) variables and five endogenous (independent) variables were employed in the model. Research indicates that a minimum of two indicators (i.e. questions) are needed to measure a construct (Kline et al., 1998; Rahim et al., 2001). In this research, the number of indicators used to measure each of the model variables ranged from two to eight, with only one variable (legislative mandates) having the minimum of two indicators. Convergent validity The measurement model of PLS analysis consisted of two phases: convergent validity followed by discriminant validity. The first phase assessed the model’s convergent validity using two measures: individual item reliability and internal consistency (Santosa et al., 2005). Individual item reliability measures the convergence of each indicator variable on its associated construct. Item reliability is assessed by examining the loading (i.e. correlations) of the indicator with their respective construct. Convergence, then, is established by comparing the loadings with a certain benchmark. All indicator loadings used in this research were above the minimum requirement (0.4) suggested by Igbaria et al. (1997) and Hair et al. (2006). The second convergent validity measure considers the internal consistency of constructs. Composite reliability was used (Fornell and Larcker, 1981) as it is considered superior to the traditional measure of consistency (Cronbach’s a) because it does not depend on the number of indicators. Adequate reliability is gained when composite reliability value is greater than 0.5. Here, all the constructs exhibited adequate reliability. Fornell and Larcker (1981) also suggest that convergent validity can be determined by using a more conservative test, which is by considering the average variance extracted (AVE). AVE should equal or exceed 0.5 to be judged adequate and again all constructs presented here meet this requirement (Table III).

Construct

Composite reliability

Cronbach’s a a

AVE

0.904 0.926 0.912 0.863 0.955 0.926 0.860 0.933 0.925 0.904

0.876 0.908 0.858 0.791 0.946 0.901 0.793 0.892 0.838 0.876

0.574 0.609 0.777 0.625 0.726 0.716 0.558 0.823 0.861 0.574


Development of indicator Managerial use of indicator Higher use of indicator Internal accountability External accountability Metric difficulties Technical knowledge Management commitment Legislative mandate Organizational capacity

Note: aPresented only for comparative purposes


Table III. Internal consistency and AVE

Discriminant validity At construct level, discriminant validity is adequate when the variance shared between a construct and any other construct in the model is less than the variance that construct shared with its indicators (Fornell, 1982; Fornell and Larcker, 1981). Regarding indicator level validity, Barclay et al. (1995) suggest that no indicator variable should load more highly on another construct than it does on the construct it is supposed to measure. Within SmartPLS, a technique known as cross loading of constructs is used to test the discriminant validity at both indicator and construct level. Table IV shows the results of the correlation of the constructs for each of the model variables. Table IV presents the correlation matrix of the construct and the square root of AVE (in italics). Previous literature recommends that these tests are best tabulated as in Table IV because the off-diagonal items (correlation of constructs) must be less than or equal to the bolded diagonal items (square root of the AVE) in the corresponding rows and columns (Barclay et al., 1995; Gefen et al., 2000). Table IV shows that diagonal values are greater than the off-diagonal values in their corresponding rows and columns, therefore there should be no issues with discriminant validity of the constructs.

Cap Cap Com Dev EAcc HUse IAcc Kno Leg MUse Met

Com

Dev

EAcc

HUse

IAcc

Kno

Leg

MUse

Met

0.758 0.390 0.907 0.238 0.581 0.758 0.491 0.614 0.483 0.852 0.224 0.490 0.501 0.424 0.881 0.275 0.620 0.442 0.642 0.644 0.790 0.339 0.516 0.513 0.528 0.361 0.466 0.747 0.189 0.505 0.510 0.491 0.361 0.507 0.515 0.928 0.179 0.732 0.652 0.545 0.608 0.552 0.556 0.546 0.781 0.073 2 0.231 2 0.280 2 0.069 2 0.109 2 0.114 2 0.172 2 0.043 2 0.252 0.846

Note: Italicised diagonal are the square root of AVE

Table IV. Correlation of constructs and the square root of AVE

PAR 24,3


276

Structural model assessment The structural model focuses on the hypothesized relationships or paths between the latent variables (Hair et al., 2006). A structural model can be used to draw conclusions about the significance of the relationship between constructs and to comment on the predictive power of the model proposed. Given the fact that SmartPLS does not assume a normal distribution of data, it is inappropriate to use traditional tests to ascertain the statistical significance between constructs (Chin and Newsted, 1999). Instead, scholars of PLS have developed two nonparametric approaches to test the relationship between variables; either jackknife or bootstrap techniques (Gefen et al., 2000; Santosa et al., 2005). In this study bootstrap was used as it is considered to be a more sophisticated approach than jackknife (Chin, 1998). It provides two measures of the structural model: a t-value (similar to the t-test) and R 2 (interpreted similarly to the traditional multiple regressions analysis). The predictive power of the research model can be assessed by using R 2 stemmed from the output of the bootstrap. Although prior research has confirmed the robustness of using PLS for non-normal data, it has still been raised by some that highly skewed data can increase bootstrap standard errors. However, findings by Hair et al. (2012), determined that reporting the extent to which data are non-normal is not often done by researchers and those studies that do show that there is very little evidence of significant differences. Table V shows the R 2 values for each of the dependent variables in the model. As indicated, R 2 values range from 0.265 to 0.607. The strongest R 2 value is that of managerial use of indicator (0.607) which indicates that 60.7 per cent of use of indicators for managerial level purposes can be explained by the constructs used in this model. The second strongest is for external accountability (0.497), while the lowest is that of higher use of indicator (0.265). All R 2 values meet the 0.10 minimum limit suggested by, for example, Santosa et al. (2005). The path coefficient results are shown in Table VI. The results range from the weakest of only 0.002 (for Met ! HUse) to the strongest of 0.529 (for Com ! MUse). Hypothesis testing To test the hypotheses, it was necessary to interpret the construct equations with standard errors and test statistics. The construct equations measure the extent to which one factor relates to another, that is, the structural path coefficient and t-values between hypothesized constructs, reflecting direct relationships (Tabachnick and Fidell, 1996). These path coefficients and associated t-values identify and demonstrate the direction and

Table V. R 2 values

Construct

R 2a

Development of indicator (Dev) Managerial use of indicator (MUse) Higher use of indicator (HUse) Internal accountability (IAcc) External accountability (EAcc)

0.455 0.607 0.265 0.437 0.497

Notes: a0.67 – substantial, 0.33 – moderate and 0.19 – weak Source: Henseler et al. (2009)


Hypothesis

Relationship

A. Development of indicator H1a Met ! Dev H2a Kno ! Dev H3a Com ! Dev H4a Leg ! Dev B. Managerial use of indicator H1b Met ! Muse H2b Kno ! Muse H3b Com ! Muse H4b Leg ! Muse C. Higher use of indicator H1c Met ! HUse H2c Kno ! HUse H3c Com ! HUse H4c Leg ! HUse D. Internal accountability H3d Com ! IAcc H4d Leg ! IAcc H5d Cap ! IAcc E. External accountability H3e Com ! EAcc H4e Leg ! EAcc H5e Cap ! EAcc

Sign.

Coeff.

t-value

– þ þ þ

20.162 0.195 0.322 0.240

1.956 * 2.224 * 2.534 * 2.735 * *

– þ þ þ

20.092 0.171 0.529 0.187

1.500 2.077 * 7.243 * * * 2.346 *

– þ þ þ

0.002 0.108 0.377 0.115

0.020 0.854 3.498 * * * 0.987

þ þ þ

0.472 0.261 0.042

3.621 * * * 2.298 * 0.346

þ þ þ

0.374 0.245 0.299

3.449 * * * 2.618 * * 2.398 *

Notes: Significant at: *p , 0.05, * *p , 0.01 and * * *p , 0.001

strength of each relationship and, as indicated, are obtained by using the bootstrapping technique in SmartPLS software. “The t-values (robust scores) need to be significant to support the hypothesized paths and should be above 1.96 or 2.56 for alpha protection level of 0.05 and 0.01, respectively” (Gefen et al., 2000, p. 35). Using this criterion, the results for the structural relationships are reported in Table VI followed by an examination of implications for the hypotheses. The discussion of the results appears in Section 7 with Figure 1 (reviewed in Section 4) providing a graphical representation of the results. Development of indicators (H1a-H4a) The results show a strong positive association between the development of indicators in ILG and performance reporting legislation (H4a, p , 0.01). In other words, as new regulations related to performance reporting emerged, ILG responded promptly by providing more indicators as required by those regulations. Further, the development of performance indicators in ILG increased as officers and staff gained more related technical knowledge offering support for H2a ( p , 0.05). The results also show support for H1a ( p , 0.05) and H3a ( p , 0.05). This indicates that metric difficulties are negatively related to the development of performance indicators and that management commitment from ILG managers is positively related to the development of indicators. Managerial use of indicators (H1b-H4b) The results reveal that the proposed relationship between management commitment and managerial use of performance indicators (H3b) is highly significant ( p , 0.001)


Table VI. Summary of hypothesis testing results

PAR 24,3


278

and therefore supported. This result clearly indicates that the use of performance indicators in ILG at managerial level is highly influenced by the commitment of ILG’s top-level management. The results also show the proposed relationships between legislative requirements, technical knowledge and managerial use of performance indicators (H4b and H2b) are both moderately significant ( p , 0.05) and hence supported. This implies that the extent of use of performance indicators by mid-level management is affected by the level of their technical knowledge and legislative requirements. The greater their level of knowledge, the more extensive the use of performance indicators. The results determine that there is no support for H1b, indicating that using performance indicators at the managerial level is not associated with the problems that exist in the process of developing performance indicators. In other words, and supporting prior literature (Mahmudi, 2003), once indicators become available, ILG’s managers just use them without a need to consider the problems that existed when they were developed. Higher use of indicators (H1c-H4c) The results reveal a strong positive association between the use of indicators at a higher level and the commitment of ILG management (H3c, p , 0.001). It is not surprising that the results provide no support for H1c. This suggests that higher level ILG managers simply use the indicators that are provided to them. As for Cavalluzzo and Ittner’s (2004) findings, this is logical, given that in government (or in this case ILG), the responsibility of developing performance indicators is usually that of senior finance officers (or equivalent). Further, results offered no support for H2c, indicating that technical knowledge is not a determinant in the use of performance indicators by higher level ILG managers. Finally, the results provided no support for H4c implying that, unlike at managerial level, the use of indicators in ILG at the higher level is unaffected by regulatory mandates received by ILG. It seems that regulations related to performance reporting (mainly imposed by central government) had no impact on level of use of indicators at a higher level. This was inconsistent with the finding in H4b. ILG managers at a higher level respond differently to the regulatory mandates than those at managerial level. Internal accountability (H3d-H5d ) The results provide strong support for H3d ( p , 0.001) and moderate support for H4d ( p , 0.05). This implies that the extent of internal accountability is influenced by both the level of commitment by ILG officers and the amount of legislative mandates imposed by central government. The results offer no support for H5d suggesting that organizational capacity (i.e. management information systems and staff capability) had no association with the extent of internal accountability practiced in ILGs. External accountability (H3e-H5e) First, similar to the results for H3d, strong support for H3e ( p , 0.001) indicated that there was a positive association between external accountability and management commitment. This implies that the extent of external accountability provided by ILGs to their stakeholders was affected by management’s commitment to implementing PMS. The results also provide support for H4e ( p , 0.01) indicating that legislation mandates had an impact on the level of external accountability of ILG. It also suggests that the mandatory nature of performance reporting regulations did ensure


improvement of external accountability of ILG. Finally, and contrary to the finding in H5d, the results provide moderate support for H5e ( p , 0.05). In other words, organizational capacity had an impact on the level of external accountability of the ILG. The higher the capacity of ILGs, the higher the level of external accountability. 7. Discussion Performance measurement PLS analysis, as shown in the previous section, provided significant results with support for many of the hypotheses. As expected, metric difficulties had a negative association with development of performance indicators, while the other three factors – technical knowledge, management commitment, and legislative mandates – had a positive significant association. These supported findings in Cavalluzzo and Ittner (2004). For example, in their study, metric difficulties caused significant problems with the extent of performance measurement development whilst their equivalent to management commitment, legislative mandates and technical knowledge had a positive association with development. Findings here also revealed that the variable legislative mandates had a strong association with development of performance indicators. This indicates the potential existence of coercive isomorphism in ILG as management tend to be more concerned about legislative requirements than other more technical (metric difficulties and technical knowledge) and normative (management commitment) factors. Bovaird and Downe (2006) found that organisations within the field of UK local government were homogenised in their adaptation of externally mandated institutional pressures. Here, there is an indication that organisations in the field of ILGs developed performance indicators more to fulfil regulatory requirements than to make their organization more effective and efficient. This is also consistent with other prior research on coercive isomorphism such as that by Arnaboldi et al. (2010) who found that the decision by government to adopt performance appraisal systems was forced on them by legislation, raising the presence of coercive isomorphism. As well, it supports our premise above that one of the main reasons for implementing PMS is to gain legitimacy rather than improve performance (Meier et al., 2006; Ashworth et al., 2009). Contrary to expectation, H1b and H1c that metric difficulties have a negative association with the use of performance indicators (both at managerial and higher level) was not supported by the results. This finding indicates that issues related to measurability did not affect the use of performance indicators by ILG managers one way or the other. Mid- and top-level managers were impervious to how indicators were developed and they only used indicators for formal reporting (legislative requirements) or political/ceremonial speeches – both reasons indicate strong support for the presence of coercive isomorphism (and also supporting prior literature relating to isomorphism and political concerns, such as de Lancer Julnes and Holzer (2001). Findings also imply that metric difficulties are more troublesome when developing indicators than when actually using them. Performance measurement was measured in two stages – development first and use second. Given this sequence, it is not surprising that the problem with measurability did not affect the use of indicators as at this stage the indicators had already been developed. Whilst support was found for H2b, that the use of indicators at managerial level was positively associated with technical knowledge, it was not the case for the use at higher level (H2c). The implication here is that issues related to technical knowledge are more


PAR 24,3


280

relevant at managerial levels than for higher levels. This may indicate that at managerial level ILG officers are more concerned with technical indicators while higher level officers normally use more macro (high level) indicators which are less technical, thus easier to comprehend. Furthermore, technical training on performance measurement for ILG is usually aimed at managerial and operational levels, not higher level managers. This, in turn, provides support for normative isomorphism. Widespread training by audit firms and universities, and the sharing of this knowledge, assist ILGs to maintain legitimacy. This dissemination of information is most common at managerial level. In line with expectations, the positive association between management commitment and use of indicators both at managerial and higher levels (H3b and H3c) was highly supported. This finding is not surprising given organizational commitment is extremely valuable. What was surprising, given Cavalluzzo and Ittner’s (2004) findings, was that use at managerial level was supported as their study only determined support for higher level use. Other literature, such as Fernandez and Rainy (2006), contends that successful change relies on participation by employees along with direction and commitment from management. de Lancer Julnes and Holzer (2001) also found that commitment has a positive impact on the use of indicators. In a country (Indonesia) that is relatively new to a decentralised form of government, commitment to use at both levels is an encouraging finding. With regard to legislative mandates, the results were mixed. Support was found for H4b that managerial use of indicators was positively associated with legislative mandates, but that was not the case for higher use (H4c). As in technical knowledge, the concern about regulation is also more relevant at managerial level than at the higher level. Again this differed from Cavalluzzo and Ittner (2004), where there was no association found for use at either level. Since the emergence of regulation on LAKIP in 1999, more technical indicators were expected to be developed and then reported in an ILG performance report. This requirement placed more pressure on managerial level officers who deal with the use of technical indicators than higher level officers who are expected to use higher level indicators that, again, are less technical. With pressure from central government to comply with legislative requirements linked to funding, coercive pressure is felt by those at managerial level. Accountability The research found that the existence of leadership was critical to the success of developing and using performance indicators in ILG. It is generally expected that the success of performance measurement implementation, in turn, will increase an organization’s internal and external accountability. Both internal and external accountability were associated with management commitment. Wang (2002) found associations between internal accountability and leadership or management commitment, however, not between external accountability and management commitment. Support here was also found for the variable legislative mandate. This indicates that one important factor that makes ILG discharge accountability is legislative requirement. The pressures from central government were still strong despite decentralization being in place for a decade. This indicates the potential existence of coercive isomorphism. With regard to organizational capacity (i.e. MIS, human resources capability), results were mixed – which was unexpected. Organisational capacity is crucial to ensure success of an organisation, including achieving accountability goals. Strong support for


the relationship between organisational capacity and accountability was consequently expected – however, this was not found to be the case here. Organisational capacity had a strong relationship with external accountability but not with internal accountability. One explanation for this inconsistency is that it does not require many resources to increase internal accountability, but rather it needs commitment from both the leader and the subordinates. Therefore, any ILG, regardless of its organisational capacity, has the potential to improve internal accountability within its organisation. Conversely, it is not an easy task for many ILGs to fulfil external accountability. Discharging accountability to external stakeholders is risky and more complex than that to internal management. Accountability can be perceived by stakeholders to exist when human resources receive further training and education. Interacting on a professional basis with others at this training coincides with normative isomorphism. 8. Conclusion From the study we can answer RQ1 that in an ILG environment there is evidence that four organizational factors (metric difficulties, technical knowledge, management commitment, and legislative requirements) have an impact on the development of performance indicators. Of these four factors, legislative requirement has the strongest effect. This implies that the main reason for developing indicators is simply to comply with central government regulation. In other words, the motivation is more about conformance than performance, and coercive pressure is strongest from central government. This, and the further suggestions of isomorphic pressures discussed below, helps answer RQ2 implying that ILGs are not using the indicators effectively. Second, metric difficulties is the only factor that has no effect on using performance indicators at managerial level. Technical knowledge through formal (e.g. training) or informal (e.g. professional group meetings) means had an impact on managerial level officers’ use of performance indicators. Normative isomorphism played a positive role in the understanding and sharing of knowledge at this level. On the other hand, management commitment had an effect on the use of performance indicators at both levels, managerial and higher level ILG managers. This was contrary to much of the prior literature and an important contribution of this research study. Third, all three independent variables tested – management commitment, legislative requirement, and organizational capacity – are evidenced to have influence on external accountability. Similar to the results achieved analysing the use of performance indicators, in regard to both internal and external accountability the strongest factor is management commitment. Institutions need to confirm their legitimacy to their various stakeholders and hence top-level managers in ILG need to give their full commitment to using PMS. If stakeholders, whether they be the citizens or central government, can see ILGs are committed to using performance indicators, funding and recognition via awards can be forthcoming. As indicated earlier, Meyer and Rowan (1977) claimed that isomorphism had three consequences for organisations; those three consequences have been found to be relevant here. First, PMS was implemented so that the ILG could be legitimated externally; second, the fact that central government based their funding and award criteria on the submission of the performance report influenced the ILGs decision to implement; and finally, implementation meant being part of an organisational field that was dependent on central government and hence this reduced “turbulence and maintained stability” (Meyer and Rowan, 1977, p. 348).


PAR 24,3


282

Academically, the model – as adapted and modified for this study – can now be used in other developing countries to analyse similar factors. As well, with its unique use of SmartPLS in the given context, this paper lays the foundation for further exploration of other theories. In saying that, in this particular environment, institutional isomorphism provided an excellent lens through which to explore and explain the possible motivations behind the development and use of PMS. Although there have been many recent suggestions (Schneiberg and Soule, 2005; Falkman and Tagesson, 2008; Lounsbury, 2008) of other ways institutional theory can be relevant, this should not detract from the work of the early authors such as Meyer and Rowan (1977) or DiMaggio and Powell (1983, 1991) and the strength of institutional isomorphism as shown here. Practically, a major contribution of this research is that it provides an understanding of factors influencing the development and use of performance measures in Indonesia and, in particular, in local government. Cavalluzzo and Ittner (2004) found that implementation of externally mandated PMS in US Government organisations was used merely to fulfil regulatory requirements, and tended to be symbolic in nature, without substantive impact on internal operations. The Indonesian Government at every tier has embraced PMS and reported performance since the reform process began. Findings here will assist local and central government, accountants, auditors, professional bodies and universities – to name a few – have more of an understanding of the factors influencing the development and use of performance measures in Indonesia. Further, they will assist with future research in other developing countries. The knowledge of these factors could, in turn, be utilised to evaluate and improve or even formulate future government policy. Local government is the government closest to the citizens and provision of quality services as measured by effective performance indicators would be an ideal outcome. Finally, to enhance external accountability more interaction and/or training could assist stakeholders (putting normative isomorphism to good use) to have more of an understanding of PMS, and accountability may not then be such an onerous requirement. Future research will consider how performance indicators are used by the various ILG stakeholders. Another area of future research is to consider why in ILG management commitment had an effect on the use of performance indicators by both levels of managers. Is ILG doing something other countries/organisations could adapt? Finally, future research will explore the development and use of PMS in other developing nations and whether alternative theories can provide as rich results as offered here. Notes 1. A list of acronyms is available in Appendix 3. 2. If you would like a copy of the survey please contact the author. References Ammons, D.N. (1995), “Overcoming the inadequacies of performance measurement in local government: the case of libraries and leisure services”, Public Administration Review, Vol. 55 No. 1, pp. 37-47. Ammons, D.N. and Rodrigues, A. (1986), “Performance appraisal practices for upper management in city governments”, Public Administration Review, September/October, pp. 460-7. Andrews, R., Boyne, G. and Walker, R. (2006), “Strategy content and organizational performance: an empirical analysis”, Public Administration Review, Vol. 66, pp. 52-63.


Arnaboldi, M., Azzone, G. and Palermo, T. (2010), “Managerial innovations in central government: not wrong, but hard to explain”, International Journal of Public Sector Management, Vol. 23 No. 1, pp. 78-93. Ashworth, R., George, B. and Delbridge, R. (2009), “Escape from the iron cage? Organizational change and isomorphic pressures in the public sector”, Journal of Public Administration Research and Theory, Vol. 19 No. 1, pp. 165-87. Barclay, D., Higgins, C. and Thompson, R. (1995), “The partial least squares (PLS) approach to causal modelling: personal computer adoption and use as an illustration”, Technology Studies (Special Issue on Researh Methodology), Vol. 2 No. 2, pp. 285-309. Barreto, I. and Baden-Fuller, C. (2006), “To conform or to perform? Mimetic behaviour, legitimacy-based groups and performance consequences”, Journal of Management Studies, Vol. 43 No. 7, pp. 1559-81. Bellamy, S. and Kluvers, R. (1995), “Program budgeting in Australian local government: a study of implementation and outcomes”, Financial Accountability & Management, Vol. 11 No. 1, pp. 39-56. Berry, A.J., Coad, A.F., Harris, E.P., Otley, D.T. and Stringer, C. (2009), “Emerging themes in management control: a review of recent literature”, The British Accounting Review, Vol. 41 No. 1, pp. 2-20. Boston, J. (2011), “Basic NPM ideas and their development”, in Christensen, T. and Laegreid, P. (Eds), The Ashgate Research Companian to NPM, Ashgate, Aldershot, pp. 17-32. Bovaird, T. and Downe, J. (2006), “N generations of reform in UK local government: compliance and resistance to institutional pressures”, International Public Management Journal, Vol. 9 No. 4, pp. 429-54. Brignall, S. and Modell, S. (2000), “An institutional perspective on performance measurement and management in the new public sector”, Management Accounting Research, Vol. 11, pp. 281-306. Cavalluzzo, K.S. and Ittner, C.D. (2004), “Implementing performance measurement innovations: evidence from government”, Accounting, Organizations and Society, Vol. 29 Nos 3/4, pp. 243-67. Chapman, C.S. and Kihn, L.-A. (2009), “Information system integration, enabling control and performance”, Accounting, Organizations and Society, Vol. 34, pp. 151-69. Cheng, R.H. (1994), “A politico-economic model of government accounting policy choice”, Research in Governmental and Nonprofit Accounting, Vol. 8, pp. 39-68. Chenhall, R.H. and Smith, K.L. (2007), “Multiple perspectives of performance measures”, European Journal of Information Systems, Vol. 25 No. 4, pp. 266-82. Cheung, A. (2011), “NPM in Asian countries”, in Christensen, T. and Laegreid, P. (Eds), The Ashgate Research Companian to NPM, Ashgate, Aldershot, pp. 131-44. Chin, W.W. (1998), “Issues and opinion on structural equation modeling”, MIS Quarterly, Vol. 22 No. 1, p. 1. Chin, W.W. and Newsted, P.R. (1999), “Structural equation modeling analysis with small samples using partial least squares”, in Hoyle, R.H. (Ed.), Statistical Strategies for Small Sample Research, Sage, Thousand Oaks, CA, pp. 307-41. Chin, W.W., Marcolin, B.L. and Newsted, P.R. (2003), “A partial least squares latent variable modeling approach for measuring interaction effects: results from a Monte Carlo simulation study and an electronic-mail emotion/adoption study”, Information Systems Research, Vol. 14 No. 2, pp. 189-217.


PAR 24,3


284

Collin, S.O.Y., Tagesson, T., Andersson, A., Cato, J. and Hansson, K. (2009), “Explaining the choice of accounting standards in municipal corporations: positive accounting theory and institutional theory as competitive or concurrent theories”, Critical Perspectives on Accounting, Vol. 20, pp. 141-74. de Lancer Julnes, P. and Holzer, M. (2001), “Promoting the utilization of performance measures in public organizations: an empirical study of factors affecting adoption and implementation”, Public Administration Review, Vol. 61 No. 6, pp. 693-708. DiMaggio, P.J. and Powell, W.W. (1983), “The iron cage revisited: institutional isomorphism and collective rationality in organizational fields”, American Sociological Review, Vol. 48, pp. 147-60. Falkman, P. and Tagesson, T. (2008), “Accrual accounting does not necessarily mean accrual accounting: factors that counteract compliance with accounting standards in Swedish municipal accounting”, Scandinavian Journal of Management, Vol. 24, pp. 271-83. Fernandez, S. and Rainey, H.G. (2006), “Managing successful organizational change in the public sector”, Public Administration Review, Vol. 66 No. 2, pp. 168-76. Field, A. (2009), Discovering Statistics Using SPSS, Sage, London. Fornell, C. (1982), A Second Generation of Multivariate Analysis, Praeger, New York, NY. Fornell, C. and Larcker, D.F. (1981), “Evaluating structural equation models with unobservable variables and measurement error”, Journal of Marketing Research, Vol. 18, pp. 39-50. Gefen, D., Straub, D.W. and Boudreau, M.C. (2000), “Structural equation modeling and regression: guidelines for research practice”, Communications of the Association for Information Systems, Vol. 4 No. 7, pp. 2-76. Gianakis, G.A. (2002), “The promise of public sector performance measurement: anodyne or placebo?”, Public Administration Quarterly, Vol. 26 No. 1, pp. 34-64. Hair, J.F., Sarstedt, M., Ringle, C. and Mena, J. (2012), “An assessment of the use of partial least squares structural equation modeling in marketing research”, Journal of the Academy of Marketing Science, Vol. 40, pp. 414-33. Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E. and Tatham, R.L. (2006), Multivariate Data Analysis, 6th ed., Prentice-Hall, Englewood Cliffs, NJ. Hall, M. (2008), “The effect of comprehensive performance measurement systems on role clarity, psychological empowerment and managerial performance”, Accounting, Organizations and Society, Vol. 33 Nos 2/3, pp. 141-63. Hood, C. (1991), “A public management for all seasons”, Public Administration, Vol. 6 No. 3, pp. 3-19. Hood, C. (1995), “New public management in the 1980s: variation and a theme”, Accounting, Organizations and Society, Vol. 20, pp. 93-109. Hood, C., James, O., Jones, G., Scott, C. and Travers, T. (1998), “Regulation inside government: where new public management meets the audit explosion”, Public Money & Management, Vol. 18 No. 2, pp. 61-8. Igbaria, M., Zinatelly, N., Cragg, P. and Cavaye, A.L.M. (1997), “Personal computing acceptance factors in small firms: a structural equation model”, MIS Quarterly, September, pp. 279-305. Indonesian Bureau of Statistics (2011), “Badan pusat statistik republik Indonesia”, available at: www.bps.go.id (accessed 3 February). Johnson, H.T. and Kaplan, R.S. (1987), Relevance Lost – The Rise and Fall of Management Accounting, Harvard Business School Press, Boston, MA.


Kaplan, R.S. (1983), “Measuring manufacturing performance: a new challenge for managerial accounting research”, The Accounting Review, Vol. 70 No. 1, pp. 71-9. Kihn, L.-A. (2010), “Performance outcomes in empirical management accounting research: recent developments and implications for future research”, International Journal of Productivity and Performance Management, Vol. 59 No. 5, pp. 468-92. Kim, S., Kim, H.K. and Lee, H. (2009), “An institutional analysis of an e-government system for anti-corruption: the case of OPEN”, Government Information Quarterly, Vol. 26 No. 1, pp. 42-50. Kline, J.P., Schwartz, G.E., Allen, J.J.B. and Dikman, Z.V. (1998), “Perceptual and electroencephalographic registration of masked emotional words in defensiveness: an exploratory study”, Personality and Individual Differences, Vol. 24 No. 4, pp. 499-512. Lapsley, I. (2008), “The NPM agenda: back to the future”, Financial Accountability & Management, Vol. 24 No. 1, pp. 77-96. Lapsley, I. and Pallot, J. (2000), “Accounting, management and organizational change: a comparative study of local government”, Management Accounting Research, Vol. 11 No. 2, pp. 213-29. Laurensius, F. and Halim, A. (2005), “Pengaruh Faktor-Faktor Rasional, Politik dan Kultur Organisasi terhadap Pemanfaatan Informasi Kinerja Instansi Pemerintah Daerah”, Proceedings of the 8th National Accounting Symposium in SNA VIII, Solo, 15-16 September, 2013, pp. 774-90, available at: http://blog.umy.ac.id/akbar/2010/12/02/ simposium-nasional-akuntansi-sna-ke-8/ (accessed 20 August 2011). LOGOTRI (2003), Performance Measurement and Management in Asian-Pasific Local Government, a Discussion Paper and Investigation into Future Directions, Centre for Local Government at the University of Technology, Sydney, Australia, 23 September. Lounsbury, M. (2008), “Institutional rationality and practice variation: new directions in the institutional analysis of practice”, Accounting, Organizations and Society, Vol. 33, pp. 349-61. McLeod, R. (2005), “Private sector lessons for public sector reform in Indonesia”, working paper, Australian National University, Canberra, 6 December. Mahmudi, A. (2003), Pengukuran Kinerja di Instansi Pemerintah Daerah, Universitas Gadjah Mada, Yogyakarta. Mahsun, M. (2005), “Analisis performance expectation gap antara public servants dengan direct users kantor pemadam kebakaran di daerah istimewa Yogyakarta”, Jurnal Akuntansi dan Keuangan Sektor Publik, Vol. 6 No. 1, pp. 156-84. Meier, K.J., O’Toole, L.J. Jr, Boyne, G.A. and Walker, R.M. (2006), “Strategic management and the performance of public organizations: testing venerable ideas against recent theories”, Journal of Public Administration Research and Theory, Vol. 17, pp. 357-77. Meyer, J.W. and Rowan, B. (1977), “Institutionalized organizations: formal structure as myth and ceremony”, American Journal of Sociology, Vol. 83, pp. 340-63. Micheli, P. and Neely, A. (2010), “Performance measurement in the public sector in England: searching for the golden thread”, Public Administration Review, Vol. 70 No. 4, pp. 591-600. Modell, S. (2004), “Performance measurement myths in the public sector: a research note”, Financial Accountability & Management, Vol. 20 No. 1, pp. 39-56. Modell, S. (2005), “Triangulation between case study and survey methods in management accounting research: an assessment of validity implications”, Management Accounting Research, Vol. 16 No. 2, pp. 231-54.


PAR 24,3


286

Neely, A. (1999), “The performance measurement revolution: why now and where next”, International Journal of Operations & Production Management, Vol. 19 No. 2, pp. 205-28. Osborne, D. and Gaebler, T. (1992), Reinventing Government: How the Entrepreneurial Spirit is Transforming the Public Sector, Addison-Wesley, Reading, MA. Pilcher, R. (2011), “Implementing IFRS in local government – institutional isomorphism as NPM goes mad?”, Local Government Studies, Vol. 37 No. 4, pp. 367-89. Pilcher, R. and Dean, G. (2009), “Consequences and costs of financial reporting compliance for local government”, European Accounting Review, Vol. 18 No. 4, pp. 725-44. Rahaman, A.S., Lawrence, S. and Roper, J. (2004), “Social and environmental reporting at the VRA: institutionalised legitimacy or legitimation crisis?”, Critical Perspectives on Accounting, Vol. 15 No. 1, pp. 35-56. Rahim, M.S., Antonioni, D. and Psenicka, C. (2001), “A structural equations model of leader power, subordinates’ styles of handling conflict, and job performance”, The International Journal of Conflict Management, Vol. 12 No. 3, pp. 191-211. Riandi, P. (2003), “Faktor-faktor Penghambat Implementasi SAKIP di Lingkungan Propinsi Kalimantan Timur”, available at: www.bpkp.go.id (accessed 11 March 2006). Ringle, C.M., Wende, S. and Will, A. (2005), “SmartPLS 2.0 (M3) beta”, available at: www. smartpls.de (accessed 28 January 2009). Romzek, B. and Dubnick, M. (1998), “Accountability”, in Shafritz, J.M. (Ed.), International Encyclopedia of Public Policy and Administration, Westview Press, New York, NY, pp. 6-11. Santosa, P.I., Wei, K.-K. and Chan, C.C. (2005), “User involvement and user satisfaction with information-seeking activity”, European Journal of Information Systems, Vol. 14 No. 4, pp. 361-70. Schneiberg, M. and Soule, S.A. (2005), “Institutionalization as a contested, multilevel process: the case of rate regulation in American Fire Insurance”, in Davis, G.F., McAdam, D., Scott, W.R. and Zald, M.N. (Eds), Social Movements and Organization Theory, Cambridge University Press, Cambridge, pp. 122-60. Scott, R.W. (1987), “The adolescence of institutional theory”, Administrative Science Quarterly, Vol. 32 No. 4, pp. 493-511. Selznick, P. (1949), TVA and the Grass Roots, University of California Press, Berkeley, CA. Selznick, P. (1996), “Institutionalism ‘old’ and ‘new’”, Administrative Science Quarterly, Vol. 4 No. 2, pp. 270-7. Shields, M.D. (1995), “An empirical analysis of firms’ implementation experiences with activity-based costing”, Journal of Management Accounting Research, Vol. 7, pp. 140-66. Silverman, D. (1971), The Theory of Organizations: A Sociological Framework, Basic Books, New York, NY. Sukarno (2006), “Akuntabilitas Kinerja: Sebuah Harapan?”, available at: www.bpkp.go.id/warta/ index.php? view¼721 (accessed 28 August). Tabachnick, B.G. and Fidell, L.S. (1996), Using Multivariate Statistics, 3rd ed., Harper & Row, New York, NY. ter Bogt, H.J. (2004), “Politicians in search of performance information? Survey research on Dutch aldermen’s use of performance information”, Financial Accountability & Management, Vol. 20 No. 3, pp. 221-52. Thornton, P. and Ocasio, W. (2008), “Institutional logics”, in Greenwood, R., Oliver, C., Sahlin, K. and Suddaby, R. (Eds), The Sage Handbook of Organizational Institutionalism, Sage, Thousand Oaks, CA.


Tsamenyi, M., Cullen, J. and Gonzalez, J.M.G. (2006), “Changes in accounting and financial information system in a Spanish electricity company: a new institutional theory analysis”, Management Accounting Research, Vol. 17 No. 4, pp. 409-32. Tuttle, B. and Dillard, J. (2007), “Beyond competition: institutional isomorphism in US accounting research”, Accounting Horizons, Vol. 21 No. 4, pp. 387-409. Usunier, J.C. (1998), International and Cross-cultural Management Research, Sage, London. van Dooren, W. (2005), “What makes organizations measure? Hypotheses on the causes and conditions for performance measurement”, Financial Accountability & Management, Vol. 21 No. 3, pp. 363-83. Wang, X.H. (2002), “Assessing administrative accountability”, American Review of Public Administration, Vol. 32 No. 3, pp. 350-70. Wang, X.H. and Berman, E. (2001), “Hypotheses about performance measurement in counties: findings from a survey”, Journal of Public Administration Research and Theory, Vol. 11 No. 3, pp. 403-28. Zucker, L.G. (1977), “The role of institutionalization in cultural persistence”, American Sociological Review, Vol. 42, pp. 726-43. Further reading de Lancer Julnes, P. (2006), “Performance measurement: an effective tool for government accountability? The debate goes on”, Evaluation, Vol. 12, p. 219. Drucker, P.F. (1999), Management: Tasks, Responsibilities, Practices, Butterworth-Heinemann, Oxford. Indonesian Government (1999), Laporan Akuntabilitas Kinerja Institusi Pemerintah/Government Institution Accountability of Performance Report (LAKIP ) – Presidential Instruction Inpres No. 7/1999, Indonesian Government, Yogyakarta, Indonesia. Wang, J. (2007), “An alternative approach to conceptualizing interviews in HRD research”, Human Resource Development Quarterly, Vol. 18 No. 2, p. 179. (The Appendices follow overleaf.)


Table AI. Constructs used in the model

5

4

3

2

Dev1: input (i.e. labour, material) Dev2: output (i.e. quantity of products or services provided) Dev3: outcome (i.e. customer satisfaction) Dev4: operating efficiency (i.e. cost/unit) Dev5: benefit (i.e. public/citizen satisfaction) Dev6: impact (i.e. achievement of social objectives) Dev7: process (i.e. narrative analysis of performance) Managerial use of Muse1: setting strategy and program priorities indicators Muse2: allocating resources Muse3: adopting new program approaches or changing work processes Muse4: coordinating program efforts with other internal or external organisations Muse5: refining program performance measures Muse6: setting new or revising existing performance goals Muse7: setting individual job expectations for government employees I manage or supervise Muse8: rewarding government employees I manage or supervise Higher use of Huse1: performance measures from my activities are used to develop my LG’s budget indicators Huse2: funding decisions for my activities are based on performance measures Huse3: changes by management above my level are based on performance measures Internal accountability IAcc1: officials at my level are held accountable for the results of their activities IAcc2: employees in my LG receive positive recognition for helping the LG accomplish strategic goals IAcc3: the individual I report to periodically reviews my activity’s results with me IAcc4: lack of incentives (e.g. reward, positive recognition) has hindered using performance information External EAcc1: organisation-wide policy priorities/goals accountability EAcc2: program goals and objectives EAcc3: program functions and activities EAcc4: program output measures EAcc5: program outcome measures EAcc6: program narrative performance information EAcc7: trends of performance measures EAcc8: comparisons of performance measures (continued)

Development of indicators

Items

288

1

No. Construct


PAR 24,3 Appendix 1

10

9

8

7

6

Legislative requirements Organisational capacity

Management commitment

Technical knowledge

Metric difficulties

No. Construct

Met1: difficulty determining meaningful measures Met2: results of our program(s)/operation(s)/project(s) occurring too far in the future to be measured Met3: difficulty distinguishing between the results produced by the program and results caused by other factors Met4: difficulty determining how to use performance information to improve the program Met5: difficulty determining how to use performance information to set new or revise existing performance goals Kno1: I receive training on development and use of performance measures Kno2: my staff receive training on development and use of performance measures Kno3: I receive published information on how to develop performance measures Kno4: my staff receive published information on how to develop performance measures Kno5: my LG involve external experts or consultants in developing performance measures Com1: my institution’s top leadership demonstrate a strong commitment to achieving results Com2: the lack of ongoing top executive commitment or support for using performance information to make program/funding decisions hindered measuring performance or using performance information? Com3: the lack of ongoing congressional commitment or support for using performance information to make program/funding decisions hindered measuring performance or using performance information? Leg1: I have been involved in my local authority’s effort in implementing LAKIP Leg2: my staff has been involved in my local authority’s effort in implementing LAKIP Cap1: management information systems Cap2: performance-based budgeting Cap3: capable staffs Cap4: budgetary surplus

Items



Table AI.

PAR 24,3

Appendix 2

Characteristics Gender

290


Age group

Education level Field background Work experience

Type Table AII. Demographic information of respondents (n ¼ 98)

Location

Female Male Missing ,30 30-40 41-50 .50 Missing Undergraduate Post-graduate Missing Accounting Others Missing ,2 2-5 6-10 11-15 .15 Missing District City In-Java Outer-Java

Frequency

Percentage

9 87 2 1 12 36 44 5 33 63 2 2 87 9 1 3 6 32 43 13 76 22 43 55

9.2 98.8 2.0 1.0 12.2 36.7 44.9 5.1 33.7 64.3 2.0 2.0 88.8 9.2 1.0 3.1 6.1 32.7 43.9 13.3 77.6 22.4 43.9 56.1

Appendix 3. List of acronyms AVE – average variance extracted ILG – Indonesian local government Inpres – Instruksi Presiden/Presidential Instruction LAKIP – Laporan Akuntabilitas Kinerja Institusi Pemerintah/Government Institution Accountability of Performance Report LOGOTRI – Local Government Training and Research Institute NGO – non-government organisation NPM – new public management OLS – ordinary least squares PLS – partial least squares PMS – perforance measurement systems SEM – structural equation modelling Research model variables – short codes Dev – development of indicator Muse – managerial use of indicator


Huse IAcc EAcc Met Kno Com Leg Cap

– – – – – – – –

higher use of indicator internal accountability external accountability metric difficulties technical knowledge management commitment legislative mandates organisational capacity

About the authors Dr Rusdi Akbar is a Lecturer in the Faculty of Economics at the Universitas Gadjah Mada in Indonesia. Dr Robyn Pilcher is an Associate Professor in the School of Accounting at Curtin University, Australia. Robyn Pilcher is the corresponding author and can be contacted at: r.pilcher@curtin. edu.au Dr Brian Perrin is a Senior Lecturer in the School of Accounting at Curtin University, Australia.

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints


This article has been cited by:


1. Rusdi Akbar, Robyn Ann Pilcher, Brian Perrin. 2015. Implementing performance measurement systems. Qualitative Research in Accounting & Management 12:1, 3-33. [Abstract] [Full Text] [PDF]

002-Akbar 2012 Performance Measurement in Indonesia-The Case of Local Government

Recommend Documents